August 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: Two silly XEmacs questions 26 years, 3 months

Adrian Aichner

>>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: sb> Speaking of 21.0, there's a new release -- 21.0-pre8 "Corsican" on the sb> FTP site that I put out this morning. I'm anticipating one or two sb> more interim releases before we put out the version for binary kit sb> building. The release announcement never materialized. Was there one? Adrian -- Adrian Aichner Applications Engineer Teradyne GmbH Semiconductor Test Group Telephone +49/89/41861(0)-208 Dingolfinger Strasse 2 Fax +49/89/41861-217 D-81673 MUENCHEN E-mail aichner(a)ecf.teradyne.com

1 participants
0 comments

Re: [PATCH] Not for 21.0. No accelerators in mswindows top-level menus 26 years, 3 months

Jan Vroonhof

Jonathan Harris <jhar(a)tardis.ed.ac.uk> writes: > menu items. The patch also causes requests for right-flushed menu items > to be ignored in mswindows frames, because right-flush is not Windows > style-compliant. Couldn't you at the very least make this configurable. Personally I would want any XEmacs to look as much the same as possible regardless of the platform I use. i.e. I prefer XEmacs style compliance over windows style compliance[1]. I do not know whether there was a lot of discussion before on the NT list but I can remember some poor XEmacs NT coder being 'coerced' by us unix folks into coding right-aligned menus for this reason so it seems silly to remove it again now. Just my $0.02 Jan [1] Adding to that: at least on the surface it seems windows style changes with every release of a new MS office version. If you were porting to the mac the situation would be different.

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Didier Verna

Gunnar Evermann <Gunnar.Evermann(a)nats.informatik.uni-hamburg.de> writes: > I guess you confuse site-load and site-start. Yup. Actually, the name (which I think should better be site-dump) has confused me for years :-)

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Gunnar Evermann

Didier Verna <verna(a)inf.enst.fr> writes: > Gunnar Evermann <Gunnar.Evermann(a)nats.informatik.uni-hamburg.de> writes: > > > I think that's what site-load.el is for (I mean dumping *packages*). I > > always wanted to create a binary with a dumped gnus -- but never found > > the time to try... > > Nope. Site-load is for ... load. Those are additional site-wide > packages to load when xemacs is started by any user. IIRW, the option -q > disables site loading. I guess you confuse site-load and site-start. from site-load.el: ;; This is a prototype site-load.el file. ;; The site-load.el mechanism is provided so XEmacs installers can easily ;; dump lisp packages with XEmacs that do not get dumped standardly.

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Didier Verna

Gunnar Evermann <Gunnar.Evermann(a)nats.informatik.uni-hamburg.de> writes: > I think that's what site-load.el is for (I mean dumping *packages*). I > always wanted to create a binary with a dumped gnus -- but never found > the time to try... Nope. Site-load is for ... load. Those are additional site-wide packages to load when xemacs is started by any user. IIRW, the option -q disables site loading.

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Gunnar Evermann

Holger Schauer <Holger.Schauer(a)gmx.de> writes: > >>>>"sb" == SL Baur schrieb am 04 Aug 1998 00:52:27 -0700: > > sb> Using the correct meaning of "dumped a package with XEmacs", > sb> nothing will happen -- you have to redump XEmacs to get the > sb> updates and in XEmacs 20.4, you will have the new Lisp package, > sb> but autoloads from the previous version, so if a global function > sb> changes, it won't be reflected in how the user sees it. > > So, is it possible in 21.0 to really dump a package or to dump > autoloads (don't know what this could be good for) ? I think that's what site-load.el is for (I mean dumping *packages*). I always wanted to create a binary with a dumped gnus -- but never found the time to try... -- Gunnar Evermann Natural Language Systems Group, University of Hamburg EMail: Gunnar.Evermann(a)nats.informatik.uni-hamburg.de

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Didier Verna

Holger Schauer <Holger.Schauer(a)gmx.de> writes: > So, is it possible in 21.0 to really dump a package Packages appearing in dumped-lisp.el are really dumped. This is restricted to some stuff from the lisp directory in the core distrib.

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Holger Schauer

>>>>"sb" == SL Baur schrieb am 04 Aug 1998 00:52:27 -0700: sb> Using the correct meaning of "dumped a package with XEmacs", sb> nothing will happen -- you have to redump XEmacs to get the sb> updates and in XEmacs 20.4, you will have the new Lisp package, sb> but autoloads from the previous version, so if a global function sb> changes, it won't be reflected in how the user sees it. So, is it possible in 21.0 to really dump a package or to dump autoloads (don't know what this could be good for) ? Holger -- --- http://www.coling.uni-freiburg.de/~schauer --- "Deine capslock und '!' Tasten klemmen." "Nee, ich habe eine Win95 Tastatur." -- Robin Socha und H.Wulff in de.comp.os.linux.misc

1 participants
0 comments

Re: Two silly XEmacs questions 26 years, 3 months

Didier Verna

SL Baur <steve(a)xemacs.org> writes: > Ah. A vocabulary gap. No no, I perfectly know what dumping and autoloading mean. I really had the idea that all packages seen at build time were dumped (yes, I mean the code). However, this was a bit silly since I know the existence of the dumped-lisp file :-/ > If you (require 'foo) where foo is a package dumped with XEmacs, Nothing > Happens as it has already been loaded. In order to get a more recent > version, you have to explicitly (load "foo"). which leads to unused code in the binary. > In versions of XEmacs prior to 21.0, the autoloads were dumped with the > binary, that is, they were static[1] from the time when the build was done. > In 21.0, the autoloads are loaded at startup time. OK. > > what happens when you have dumped a package with xemacs, and you > > install a new version without building xemacs again ? > > In XEmacs 21, you will have the new package as if the first version of > the package *never existed*. This is a huge win. I hope more folks > can understand this. OK.

1 participants
0 comments

[ EXPERIMENTAL PATCH ] package-get mods for local files (CDROMS) 26 years, 3 months

Darryl Okahata

[ This is *not* being sent to xemacs-patch, as I'd like to get some feedback and maybe fix up some things. A ChangeLog is not included because this patch is still experimental. ] This patch fixes the package-get routines to optionally look on a local disk for packages. Basically, package-get-remote was modified to allow for a system name of `nil', in which case the directory is taken to be a local directory. This allows the user to set package-get-remote in his .emacs file (for example): (setq package-get-remote '( (nil "/my/local/directory") (nil "~/local/package/dir") ; note: expansion is done ("ftp.xemacs.org" "/pub/xemacs/beta/xemacs-21.0/packages/binary-packages") )) This causes `package-get' to look in the local directories, "/my/local/directory" and "~/local/package/dir", for packages. This is a big win for mswindows, as the user can then add a pathname to a CDROM containing all of the packages. I also enhanced most of the interactive functions to do package completion; whenever you're queried for a package name, you can now do completion. Also, for mswindows users, `package-admin-add-binary-package' in package-admin.el was modified to indirectly call an installation function, to actually do the work of installing the package: ------------------------------------------------------------------------------- (defvar package-admin-install-function 'package-admin-default-install-function "The function to call to install a package. Three args are passed: FILENAME PKG-DIR BUF Install package FILENAME into directory PKG-DIR, with any messages output to buffer BUF.") ------------------------------------------------------------------------------- The default function preserves the existing functionality of `package-admin-add-binary-package' (i.e., it calls "add-big-package.sh" to do the actual work). The mswindows users can now set `package-admin-default-install-function' to `package-admin-install-function-mswindows', to install packages. However, this function uses "djtar" to extract .tar.gz packages, and few people have djtar (not to mention that cygwin users probably want to use the Unix script). Perhaps this function should be deleted? The patch is attached. Note that it's relative to the xemacs-20/lisp directory (sorry about that). Issues: * I'm thinking that this is a 21.1 feature, as subprocesses are broken on native Win95. It works under WinNT, as long as efs doesn't get into the act. * mswindows users need to be able to extract .tar.gz files. I'm using djtar (a native Win32 version, and not a djgpp-compiled one); should the code be using something better? * What about cygwin folks? * Should `package-admin-install-function' be dynamically set to use the correct package extraction function? Right now, it's hard-coded to `package-admin-default-install-function', but this is wrong for mswindows users. What do cygwin folks want? * Should I, or should I not submit this to xemacs-patches (properly cleaned up, of course)? Comments? -- Darryl Okahata Internet: darrylo(a)sr.hp.com DISCLAIMER: this message is the author's personal opinion and does not constitute the support, opinion, or policy of Hewlett-Packard, or of the little green men that have been following him all day.

3 participants
4 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta August 1998